Because glacier ice is a relatively well characterized material, its
behavior is generally predictable. As recognized by Nye (1952, Journal of Glaciology,
2, 82-93 and 103-107) and discussed by others in the intervening years, this
predictable rheology can be used to reconstruct former glaciers. Only minimal field
data are required - specifically, terminal moraines and valley geometry. Alternatively, if
the vertical extents of former glaciers can be reconstructed on the basis of field
evidence (moraine crests, trimlines, nunataks, etc.), flow models can be used to
interpret variability in the factors which control ice flow, especially effective basal
shear strength. The discussion below involves the application of iterative models to
former ice cover on the northern Rocky Mountains of Montana. Companion pages discuss
the former ice cover of other mountain masses of the western United States and their
interpretation.Northern Rocky MountainsI reconstructed the former ice cover over the northern Rocky Mountains of Montana (Figure 1), south of the Canadian border and the Cordilleran Ice Sheet (1995, Journal of Geomorphology, 14, 123-130). Contours are shown at 500-m increments. Elevations over 2500 m are shaded. The major drainages and lakes are shown in blue. Scale bar is 50 km. |
 |
| I used a spreadsheet model to reconstruct centerline elevations along major flowlines (major streams in Figure 2). Starting positions were defined by major, previously-mapped, moraine systems (heavy dashed lines). Within the mountains the only constraints were the elevations of breached divides (red ")(" symbols) and nunataks (green dots), both interpreted from topographic maps on the basis of presence or lack of obvious glacial scouring and shaping. Low-elevation effective basal shear stresses of 10-25 kPa (0.1 - 0.25 bar), well-constrained by the moraines, generally implied strongly extending flow and probably deforming bed material as well. However, an assumed 100 kPa (1 bar) effective basal shear stress above the terminal lobes obeys common observations of glacial behavior and fit the morphological constraints (breached divides and nunataks) well. |  |
| The reconstructed ice cap (Figure 3), despite the strong NNW-SSE
structural grain imparted by Thrust Belt deformation and evident in Figures 1 and 2, shows
a central dome exceeding 2500 m in elevation defined by solid 500-m master contours and
dashed 100-m supplemental contours. Ice from that dome apparently drained NE into the Two
Medicine Lobe, WNW (then W and S) into the Flathead Valley, and radially outward into
smaller distributary outlet lobes (Sun River to the E; Blackfoot and Clearwater to the SW)
and glaciers. This reconstruction may be testable based on the distribution of erratic
cobbles from distinctive units in the thrust sheets. Personal communications from regional
field geologists suggests that this model underrepresents the contribution of Cordilleran
ice to the Two Medicine Lobe. However, how ice could have drained from the NW, across the
divide at Marias Pass (W of the large Two Medicine lobe) and onto the plains, rather than
directly into the Flathead Lobe, is unclear. The value of this study is several-fold:
|
 |